Latest News

November 20, 2019

Excel Spreadsheet Analytics vs. SaaS Analytics

When does is make sense to upgrade your analytics?

Most corporations use Excel spreadsheets for a variety of business requirements, and for good reason. Excel is inexpensive, easy to learn, easy to customize and has useful computational features. However, as companies and their analytic needs grow, the strengths of Excel can quickly become its faults. Excel spreadsheet flexibility and customizability can enable unseen mistakes and financial errors, especially when being used by multiple individuals.

Software-as-a-service (SaaS) analytics, as compared to Excel analytics, are not only more robust, but are also highly secure and scalable. SaaS typically means a software solution that is cloud-based and accessible via a web browser.  SaaS analytics are generally superior to local (downloaded) software as they eliminate any associated hardware, costly implementation headaches and IT costs. Below, we summarize specifics as to why SaaS analytics, in some cases, should replace Excel spreadsheets.

Spreadsheets are vulnerable to human error.

The flexibility and customizability of Excel spreadsheets makes them easy to use.  However, these characteristics also make spreadsheets more likely to contain errors as complexity increases. The use of inter-cell and inter-tab formulas makes error tracking time-consuming and difficult within spreadsheets. With every cell subjected to manual manipulation, one miscalculation can compromise further calculations and results. This article, “The 7 Biggest Excel Mistakes of All Time”, describes common human errors in corporate spreadsheets which resulted in millions of dollars in losses for those organizations.   

SaaS analytics are less susceptible to these errors since typical end-users of SaaS analytics do not adjust the formulas or the underlying analytical code.  There are three main steps in order to run a SaaS analytics model:

  1. Company data is uploaded
  2. Model is run
  3. Results are downloaded

The analytic models are usually hosted in a cloud-based platform, significantly decreasing the risk of introducing unforeseen human errors. Model formulas can be adjusted, but this is reserved for specific personnel trained for that purpose.

SaaS analytics facilitates collaboration and control.

The fastest and most common way to share spreadsheets internally is through email.  Not only does this compromise data security, but it also can also make keeping track of multiple file versions and edits difficult. Here are a few common and potentially damaging mistakes:

  • Sally sends John a spreadsheet.  John makes 10 edits but only tells Sally about 7 changes. How does Sally find and validate the other changes?
  • Sally is updating the corporate spreadsheet and has asked other team members to review different tabs.  As the other team members make changes, there are several versions of the spreadsheet in circulation.  Bringing all changes together into a single updated spreadsheet is time-consuming and error prone. 
  • Jenny accidentally mis-titled a spreadsheet. Ben needs to share this spreadsheet with the accountant but sends the wrong version resulting in several days of confusion.
  • John has been managing the corporate spreadsheet for years and suddenly retires.   A new analyst may take weeks to fully understand John’s complex spreadsheet.  Often the new analyst finds mistakes that John had missed for years.

SaaS analytics can solve these problems.  These analytic tools are often accessed within a standard, easy to use platform that is accessed in a web-browser.  Multiple users or groups can use these tools simultaneously without the need to share models via email.  An intuitive user interface means that new or junior staff can successfully manage complex analytic processes.  Finally, if model changes are required, this can be managed in a systematic way.  Spreadsheet fans will point out that incorporating new changes may require days for SaaS analytics, rather than minutes for a spreadsheet.  That is true.  But if your corporate model is mission-critical, you probably want to manage these changes more systematically anyway. 

SaaS analytics are more secure.

Excel does allow users to password protect many elements, hide formulas and lock cells.  When using spreadsheets this is a good first line of defense but does not solve many underlying issues.  Complex spreadsheets are still prone to human error and locking or hiding those errors compounds the problem.  Also, there are many ‘un-protect’ programs available to unlock a spreadsheet and gain access to its data.  If a hacker does get a hold of a confidential spreadsheet, Excel can’t stop them from copying information and pasting it to another workbook.

SaaS analytics are able to leverage many cloud security features to ensure that sensitive data remains secure.  At cQuant.io, we use numerous security features in our SaaS platform that ensure our customer’s data is protected. These features include:

  • Secure communication (HTTPS)
  • Encryption of user and company data
  • 2-factor authentication
  • User session time outs/lock outs
  • Password complexity & change management

Like many SaaS providers, cQuant.io deals with sensitive customer data every day and works hard to ensure that we are employing the latest security tools and protocols.

SaaS analytics provides a more robust workflow.

A large downside to Excel is that if calculations are complex enough, Excel has difficulty managing the workflow.  Some corporate spreadsheets have become so large that they take tens of minutes to open and hours to calculate when new data is introduced.  This is due to Microsoft Excel’s software being relatively slow, even when the calculations are simplistic.    

For most spreadsheet models, the computations are single-threaded and deterministic.  Most companies would benefit from seeing scenarios and simulations of future outcomes.  This is especially true in energy and commodities organizations where market prices and weather are key drivers of risk and are quite volatile.   The company CFO and/or Risk Manager may like to see a “gross-margin-at-risk” report which provides an expected gross margin with useful distributions around the mean.  However, this is very difficult to enable and manage in Excel, so the risk analysis is often not performed.   

In the analytics world we call this “model limited choice”.  Your spreadsheet model cannot manage a robust analytical process, so that process is simply not performed. In this case, the company has less information on which to make decisions because of its choice of model. 

SaaS analytics are typically built using robust statistical analysis software. cQuant.io’s models are built in R, Python, C++, and other high-performance programming languages.  These languages are extremely fast, flexible and powerful.   The models are managed by our cloud platform, providing many user tools and data visualizations.  Multiple users can run the same models at the same time and share results.  cQuant’s analytics also preserve your workflow through time. Because cQuant’s platform saves every model run, the user(s) can look back to historical runs, model settings, and results.

Our Conclusion

Overall, there are many upsides to using Excel spreadsheets and not all corporate models should be converted to SaaS analytics. But organizations should consider converting spreadsheets to SaaS analytics when those models become more complex, mission-critical, multi-user and especially when mistakes can cost the company millions.

Are you ready to improve your energy analytics? Request your free demo today.

SCHEDULE YOUR DEMO TODAY

Scroll to top